Difference between CMUDict of None?

NVIDIA / mellotron

Mellotron: a multispeaker voice synthesis model based on Tacotron 2 GST that can make a voice emote and sing without emotive or singing training data

BSD 3-Clause "New" or "Revised" License

855 stars 183 forks source link

Open lqniunjunlper opened 4 years ago

lqniunjunlper commented 4 years ago

Is this have a great impact for generated audio quality between using CMUDict or not?

rafaelvalle commented 4 years ago

Using a phoneme dictionary can be helpful when using languages like English that are not phonetic.